Inside Hex: How CMO Ashley Stepien Is Rebuilding Marketing with AI

In this episode of AI’ve Got Questions, Stacey sits down with Ashley Stepien, CMO of Hex, to talk about how AI is transforming modern marketing. Ashley shares how her team uses tools like ChatGPT and Claude to automate tedious tasks, test messaging at scale, and even rethink the structure of a marketing org.

They dive into the balance between build vs. buy in the age of AI, what it means to create “AI-native” marketers, and why brand creativity is making a big comeback.

If you’re a marketing leader trying to figure out where to start, or how to go deeper, with AI, this one’s packed with practical inspiration.

Transcript

AI’ve Got Questions — Season 2, Episode 1

Guest: Ashley Stepien, CMO at Hex
Host: Stacey Epstein

Stacey Epstein (00:29)
Today on the podcast, as we kick off our new CMO-focused season, I’m thrilled to have my friend Ashley Stepien joining me. Thanks for being here, Ashley.

Ashley Stepien (00:39)
Happy New Year! Thanks for having me.

Stacey Epstein (00:40)
As I thought about evolving the podcast to hear from real-world CMOs about how they’re using AI, you were literally the first person who came to mind.

We were at that dinner a few weeks ago, and you had so many interesting examples of how you’ve evolved your marketing with AI innovations. I’m excited to dig in—some of what we talked about, I want to hear how it’s going.

But before that, let’s introduce you properly. You’ve been at some amazing companies—tell us a bit about your journey and how you got to where you are today.

Ashley Stepien (01:24)
Sure. I actually came up through the SDR route. My first job out of college was pounding phones, just trying to make a buck. I share that because it gave me a lot of empathy for sales and how to partner better with that team to drive revenue.

At the end of the day, marketing’s job is to drive growth and revenue, and that early experience really shaped how I think about it. I’ve worked at companies like Marketo, Pendo, Webflow, then moved to Ramp, Amplitude, and now Hex.

It’s been a great run—I’ve seen so many changes, but honestly, I feel like I’ve learned more in the last six months than in the previous six years. It’s an incredible time to be a marketing leader.

Stacey Epstein (02:25)
It really is—exciting and maybe a little intimidating too.

I love that you started in sales. I did too, and it gives such perspective—you’re on the front lines, trying to make the positioning land and close a deal. It either works or it doesn’t. That appreciation for sales makes for stronger marketing leaders.

But back to this “crazy time.” When I get podcast guest requests, a lot of marketers are still just trying to figure out where to start with AI. We’re a couple of years into this Gen AI revolution, and yet many teams haven’t embraced it beyond using ChatGPT here and there.

So let’s start simple: what were the very first things you did to leverage AI as a marketing leader—and then how has that evolved?

Ashley Stepien (03:37)
I think most people have at least experimented with ChatGPT or Claude. And if you haven’t, that’s still a fair starting point. The power of these tools can’t be underestimated.

I heard someone say, “You’ll get dumber if you use ChatGPT for the wrong things—but smarter if you use it for the right ones.” That really resonated.

If you just ask it to write or summarize for you, that’s probably the “dumber” category. But if you use it as a thought partner—to help solve a critical problem—that’s where you get smarter.

So my first tip: challenge these tools to think with you, not for you.

The second: every tool in your martech stack is now adding AI. Salesforce, Notion, Webflow—whatever you use, explore the AI features that are already there. That’s low-hanging fruit and fits naturally into existing workflows.

Stacey Epstein (05:46)
So good. That second one especially—every company is racing to integrate AI, so just start there.

Let’s go back to ChatGPT for a moment. Can you share some real examples of how your team has used it?

Ashley Stepien (06:15)
Yes! The big aha moment for me was realizing we could use tools like Claude to build what it calls “artifacts”—essentially mini products that live on and can be reused.

Instead of using ChatGPT or Claude just as a conversation, you can have them create these reusable tools.

We got the team together and listed everything we hated doing. The mundane stuff and the hard stuff.

Example 1 – Mundane: We hated A/B-testing emails. It felt like such a time suck. So we asked Claude to build an “email testing engine.” We gave it a prompt describing what we wanted to test—audience, subject line, time of day, geo, content, CTA, sender—and it built a little tool where we could drop in two or three email versions. It automatically tested and told us which one to send.

That workflow went from 45 minutes to five.

Example 2 – Hard: Product marketing. PMM is subjective—everyone has an opinion. We wanted to make it more objective, especially around testing messaging and positioning.

We built a synthetic dataset with Claude. We defined our ICP—roles, company sizes, industries, geos—and had it create a representative audience of personas (“Jenny, Chief Data Officer at a 500-person company,” etc.). Then we built a scorecard to test our messaging across criteria like urgency, relevance, differentiation, clarity.

We’d drop our elevator pitch into the artifact, and it would simulate how each persona would rate it.

We did in two hours what used to take months with research firms. It gave us quantitative feedback—what resonated by role and company type—and we could optimize instantly.

Stacey Epstein (11:25)
That’s incredible. And I was thinking of your launch video—congrats on the new launch!—because I just watched it. When you’re creating something like that, everyone has opinions: CEO, head of sales, etc. Now you can actually test it and see what will resonate with your audience, not just internally.

Ashley Stepien (12:24)
Exactly.

Stacey Epstein (12:27)
That’s the beauty of AI. Things that once took research firms and months of work can now be done in hours.

My motto with AI has become: whenever we have something to do, my default is “How can AI make this faster and better?”

And I love your idea of gathering the team to rank the things nobody likes to do—and just ask AI to fix those first.

Ashley Stepien (13:21)
Totally. And that leads to the build-versus-buy question. We’ve been asking that forever in software, but it’s especially interesting in the age of AI.

In many cases, I can build to 80 percent of what I could buy. So I’m challenging vendors to think bigger—don’t just solve one small use case, think about how you can empower the whole marketing function. It’s going to be a fascinating next 12 months as the real value shakes out.

Stacey Epstein (14:45)
That makes total sense. It also raises a question about team structure. Do you have someone dedicated to experimenting with AI—or is everyone expected to dive in?

Ashley Stepien (15:01)
We’re building a culture where everyone experiments, but I do have someone who leads the charge. She’s technical, data-oriented, naturally curious—comes from operations.

But I’ve told her: don’t do all the AI work yourself. Teach everyone else to do it. Be the AI czar, not the AI bottleneck. The goal is to help everyone learn to fish, not keep feeding them.

Stacey Epstein (16:07)
I love that. And you’re right—it’s not that hard.

Honestly, I built this podcast through AI. I literally asked, “How do I start a podcast?” and it told me: record on Riverside, host on Buzzsprout, and so on. The curiosity starts there.

Even for personal stuff—like planning my daughter’s bat mitzvah—when I need to merge lists or figure out a spreadsheet formula, I just ask AI how to do it.

That “go figure it out in AI” culture is powerful.

Ashley Stepien (17:30)
Exactly. It’s the greatest gift we can give our teams right now.

My team is mostly junior, by design, and I think our job as leaders is to help them become the marketers the industry will need one or two years from now. The makeup of marketing orgs is changing fast, and we owe it to them to guide that transition.

Stacey Epstein (18:28)
Speaking of teams—you mentioned a pretty major restructuring. Tell us what you changed and how it’s going.

Ashley Stepien (18:52)
It’s evolving daily, but a few things are clear.

First, marketing orgs are going to flatten. I think the “dot-connectors”—the middle managers who coordinate everything—will become less necessary. There’ll be two camps: builders and connectors. As tools become more powerful, builders will need fewer connectors.

Second, specialization will fade. We’ll still need core functions like demand gen, product marketing, and brand—but individuals will become more well-rounded. I’m pushing my team to become generalists, almost like mini-GMs.

Instead of giving them a job, I give them a problem: “Go win the enterprise in California.” Traditionally, that would require demand gen, field marketing, PMM, comms, brand—all hands on deck. Now, with AI, one person or a small pod can self-serve much of that.

So I’m encouraging everyone to:
 1️⃣ Learn disciplines outside their specialty, and
 2️⃣ Unlearn habits that no longer serve them.

First-principles thinking means starting from scratch and questioning everything, which is hard—especially for experienced marketers—but essential in this new era.

Stacey Epstein (22:22)
Exactly. The ultimate vision is a team of multifunctional builders using AI to help craft positioning, segmentation, and campaigns—all with the help of their AI copilot.

We CMOs already have to be generalists, and now AI makes that possible for everyone. It’s such fresh thinking—most people just try to apply AI to what they already do, but you’re reimagining how we work.

That’s where the real transformation is happening.

Ashley Stepien (24:37)
Wish me luck!

Stacey Epstein (24:40)
You got it! Any last thoughts before we wrap?

Ashley Stepien (24:48)
Yes—this era of marketing is fascinating. I think we’re about to see a big return to the “arts and crafts” side of marketing. For years we fought to prove we were strategic business drivers. But now, brand and creativity are becoming the real differentiators again.

AI has freed us from so much manual work—it’s time to reclaim the fun, creative part of marketing.

Stacey Epstein (25:51)
I love that. I got so tired of defending my tiny brand budget, so it’s great to see creativity making a comeback.

AI can handle the execution, but the human creativity—that spark—is what truly differentiates us.

Well, Ashley, this has been such a great conversation. I can’t wait to check back in a few months and hear how the new team model is evolving. Thanks again for joining us.

Ashley Stepien (26:39)
Thanks for having me. Looking forward to it.

Stacey Epstein (27:06)
Thank you.